EP1826724A1 - Edition d'image au niveau objet utilisant les titres des données d'image - Google Patents
Edition d'image au niveau objet utilisant les titres des données d'image Download PDFInfo
- Publication number
- EP1826724A1 EP1826724A1 EP06270025A EP06270025A EP1826724A1 EP 1826724 A1 EP1826724 A1 EP 1826724A1 EP 06270025 A EP06270025 A EP 06270025A EP 06270025 A EP06270025 A EP 06270025A EP 1826724 A1 EP1826724 A1 EP 1826724A1
- Authority
- EP
- European Patent Office
- Prior art keywords
- image
- level information
- source
- information
- objects
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000000034 method Methods 0.000 claims abstract description 74
- 238000012545 processing Methods 0.000 claims description 12
- 230000011218 segmentation Effects 0.000 claims description 12
- 230000008569 process Effects 0.000 claims description 11
- 238000005520 cutting process Methods 0.000 claims description 8
- 238000004590 computer program Methods 0.000 claims description 6
- 230000008921 facial expression Effects 0.000 claims description 4
- 238000005286 illumination Methods 0.000 claims description 3
- 244000025254 Cannabis sativa Species 0.000 abstract description 16
- 230000006870 function Effects 0.000 abstract description 9
- 230000006872 improvement Effects 0.000 abstract description 4
- 238000010586 diagram Methods 0.000 description 25
- 238000009826 distribution Methods 0.000 description 7
- 241000283690 Bos taurus Species 0.000 description 6
- 230000009471 action Effects 0.000 description 6
- 241000593989 Scardinius erythrophthalmus Species 0.000 description 4
- 238000004458 analytical method Methods 0.000 description 4
- 239000000463 material Substances 0.000 description 4
- 238000002156 mixing Methods 0.000 description 4
- 230000015572 biosynthetic process Effects 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000003860 storage Methods 0.000 description 2
- 238000003786 synthesis reaction Methods 0.000 description 2
- 241001408627 Agriopis marginaria Species 0.000 description 1
- 241001465754 Metazoa Species 0.000 description 1
- 241001494479 Pecora Species 0.000 description 1
- XUIMIQQOPSSXEZ-UHFFFAOYSA-N Silicon Chemical compound [Si] XUIMIQQOPSSXEZ-UHFFFAOYSA-N 0.000 description 1
- 239000003086 colorant Substances 0.000 description 1
- 239000002131 composite material Substances 0.000 description 1
- 238000007796 conventional method Methods 0.000 description 1
- 238000012937 correction Methods 0.000 description 1
- 230000000717 retained effect Effects 0.000 description 1
- 238000004513 sizing Methods 0.000 description 1
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T11/00—2D [Two Dimensional] image generation
- G06T11/60—Editing figures and text; Combining figures or text
Definitions
- This description relates generally to image editing; it is particularly related to, but in no way limited to, systems and methods for editing digital images using information about objects in those images.
- Image processing tools such as Microsoft's Digital Image Suite (trade mark), Adobe Photoshop (trade mark) and Illustrator (trade mark) as well as Google Picasa (trade mark) are known.
- Microsoft's Digital Image Suite trade mark
- Adobe Photoshop trade mark
- Illustrator trade mark
- Google Picasa trademark
- it is required to provide increased functionality within the tools such that users can perform more types of operations on digital images as well as to improve existing functionality such that the results of operations are more realistic and aesthetically pleasing.
- the information about objects comprises a segmentation, depth ordering information and/or information about the class each object is a member of.
- classes include sky, building, aeroplane, grass and person.
- This object-level information is used to provide new and/or improved editing functions such as cut and paste, filling-in image regions using tiles or patchworks, digital tapestry, super resolution, auto cropping, auto colour balance, object selection, depth of field manipulation, object replacement, and object synthesis.
- improvements to user interfaces for image editing systems are described which use object-level information.
- the present example provides a method of editing a source digital image comprising the steps of:
- Another example provides a corresponding apparatus for editing a source digital image comprising:
- the object-level information comprises information about depth ordering of at least some of the objects.
- it also comprises object attributes and positions of the object parts.
- the method further comprises combining at least part of the source digital image with one or more reference digital images to form the output digital image and wherein the method comprises accessing automatically generated object-level information about the reference digital images, that information comprising segmentation of the reference images into one or more objects and, for each of the objects, details of a class that the object is a member of.
- the step of automatically processing the source image and any reference images comprises using tiles and selection of the tiles is carried out on the basis of the object-level information.
- the tiles are to be added to a particular object in the source image or reference images, those tiles being selected on the basis of the object-level information about that particular object, the selection being from the particular object or object part in any of: the source and reference images, other objects of the same class in the source and reference images, and a store of other objects of the same class.
- the tiles are selected on the basis of depth ordering information.
- step (ii) of automatically processing the image comprises cutting or copying an object from the source image and pasting it into at least one target image; said method further comprising accessing object-level information about the target image, and pasting the object into the target image on the basis of the object level information about the target image to form the output image.
- method further comprises cutting the object from the source image and filling in any regions of that object that had been occluded in the source image on the basis of object-level information about the object.
- the method further comprises pasting the cut or copied object into the target image after automatically editing the cut or copied object to be pasted on the basis of the object-level information about the target image where said object-level information can comprise semantic information.
- the step of automatically editing the cut or copied object comprises any of: resizing, recolouring, refocusing, changing the orientation, changing facial expressions; and changing illumination.
- Another example provides a computer program comprising computer program code means adapted to perform all the steps of any of the methods described above when said program is run on a computer.
- the computer program can be embodied on a computer readable medium.
- the method may be performed by software in machine readable form on a storage medium.
- the software can be suitable for execution on a parallel processor or a serial processor such that the method steps may be carried out in any suitable order, or simultaneously.
- FIG. 1 is a schematic diagram showing inputs to an object-level image editing system.
- a source image 10 of any suitable type in digital form is provided to an object recognition system 11.
- the source image 10 in the present example shows a light aircraft standing on grass in front of a building. A cloudy sky is visible above the building.
- the object recognition system produces an analysis of the source image.
- the analysis results are represented in FIG. 1 as an image 12 although this is not essential. Any suitable form of representation can be used.
- the analysis of the source image comprises a segmentation of that source image into one or more objects. In the example shown the image is segmented into four main regions which are indicated by shaded regions in image 12 of FIG. 1.
- the object recognition system also labels each segmented region as being of a particular type of object class.
- the four regions are identified as being of object classes sky 14, grass 16, building 1 5 and plane 17.
- the information produced by the object recognition system is referred to herein as object-level information and this is accessible to the image editing system 1 3 of the present invention together with the source image 10 itself.
- any suitable type of object recognition system 11 can be used which has the ability to segment a source image and identify appropriate object classes for segmented regions of the source image. At least a plurality of possible object classes is used to label the segmented image regions and in a preferred example, around 20 object classes it is possible to increase the scale of the system and use much higher numbers of object classes. Examples of object classes comprise sky, grass, person, building, horse, plane, cow, sheep, water, boat, tree, road and flower.
- the object recognition system 11 is preferably also able to provide depth ordering information for the identified objects although this is not essential.
- the object recognition system is preferably, but not essentially, able to identify particular instances of objects (e.g.
- the object recognition system may optionally comprise semantic information associated with the object classes such as information about the relative sizes of objects of class person and building or information about the general position of objects such as sky in images.
- FIG. 2 is a schematic diagram of the object-level image editing system 13 of FIG. 1 in more detail.
- One or more source images 10 and optionally one or more reference images 9 are input to the object recognition system 11 discussed above.
- the source image may comprise a photograph of a person against a walled background.
- the reference image may comprise a garden landscape.
- the output image would comprise an image of the person against a garden landscape.
- the object recognition system is preferably used to analyse both the source images and reference images. The recognition results, in addition to the source 10 and reference images 9 are made available to the object-level image editing system 13.
- That system comprises an object-level information interface 8 which interfaces with the object recognition system 11 or accesses results of that object recognition system in any suitable manner.
- a processor 7 is also provided and optionally a user interface 6.
- the processor is of any suitable type such as a personal computer, an operating system, a dedicated processor or a distributed system.
- the object-level image editing system produces an output image 5 on the basis of the various inputs.
- any of the embodiments described herein it is also possible for a user to optionally modify or correct the object-level information prior to using that information for image editing.
- the user can make these modifications or corrections manually or in any other suitable manner.
- FIG. 3 is a high level flow diagram of a method of editing an image using an object level image editing system.
- Object level information about one or more source images, and optionally one or more reference images is accessed (see box 35). This is done using the interface 8 of FIG. 2.
- the source image(s) and any reference image(s) are processed on the basis of the object level information (see box 36) and an output image formed (see box 37).
- the example-based filling system described identifies appropriate filling material to replace a destination region in an image and fills in the destination region using this material, thereby alleviating or minimising the amount of manual editing required to fill in a region in an image.
- Destination regions may be designated by user input (e.g. selection of an image region by a user) or by other means (e.g. specification of a colour or feature to be replaced).
- tile is used herein to refer to a region or patch of image data suitable for filling a destination region of a digital image.
- a tile can be represented by a group or grid of pixels or in any other suitable manner.
- FIG. 4 is a flow diagram of a method of using tiles in the object-level image editing system of the present application.
- the tiles of image data are patches of pixels or grids of pixels.
- the tiles of image data can be used for example, to fill in regions of an image where an object has been deleted.
- the tiles of image data are selected on the basis of the object-level information (see box 38). For example, they are selected from an object of the same class (or the same object part) as an object/parts which needs to be filled.
- the tiles can be selected from any of: the source and reference images, other objects of the same class in the source and reference images, and a store of other objects of the same class (as the object being filled).
- FIG. 5 gives two example user interface displays 35, 36 from our object-level image editing system used for semantic object removal.
- Display 35 comprises an image of a person 31 standing in a grassed area with a path receding into a background 30 which has trees and sky visible.
- Each display 35, 36 has graphical user interface devices such as a hierarchical selection structure 34 and an operation bar 33.
- the operation bar 33 indicates that the focus of the image is being worked on, presents a slider bar 33 to adjust the amount of blur in the image and provides a selection box to make the image greyscale rather than colour for example. Buttons to apply the settings or cancel them are also provided.
- the user has selected the person 31 in any suitable way, such as by placing the mouse pointer over the person 31 and is presented with a menu 32 of available actions including in this example, erase, focus, fix red eye, and colour levels. Suppose the user selects the erase function from menu 32.
- the image of the person 31 is "semantically" erased giving the result shown in display 36.
- the image region previously taken up by the person is indicated by reference numerals 40 and 41 in FIG. 5.
- Depth ordering and shape information is used to infer that the region behind the person is to be filled in using "grass like” tiles in region 41 and "tree like” tiles in region 40. Without this depth ordering information it could be that "grass like” tiles are used to fill in the whole region where the image of the person 31 previously was. Also, if there were objects in front of the person being erased, the depth information would allow them to be ignored during the fill in.
- segmentation information, segmenting the image into grass, trees, people objects and sky is used to infer which types of tiles to use.
- the processes of copying and cutting image regions are preferably identical except that the process of cutting also involves filling in the area of source image from which a region is cut.
- the filling in process is carried out intelligently using object-level information.
- FIG. 6 is a flow diagram of a method of using the object-level information for smart copy and paste, or smart cut and paste.
- a source image comprises an object to be cut or copied and a reference image which in this case is referred to as a target image is provided to paste the cut/copied object into.
- the method comprises accessing object-level information about the source and target image as discussed above (see box 60 of FIG. 6).
- the object-level information is used to copy or cut an object from the source image (see box 61).
- the segmentation information is used to identify objects and one of these is selected either automatically (for example, using depth ordering information to pick a foreground object, or using pre-specified rules in conjunction with the object class information) or as a result of user input.
- any suitable method of cutting or copying that object is used as known in the art.
- we preferably also fill-in the area of the source image from which a region is cut. This filling in is carried out intelligently using object-level information.
- An assessment is then made as to whether there are any regions of the cut/copied object which are occluded in the source image. Depth ordering information and or other object-level information is used to make this assessment. If there are occluded regions, these are filled in (see box 63) using tiles selected using the object-level information from the source image (as discussed above). In addition regions in the source image that are left blank as a result of any cut operation are filled in if required. This is done using tiles selecting using the object-level information.
- the next stage comprises comparing the object-level information about both the cut/copied object and the target image and deciding whether the cut/copied object needs to be edited (see box 64).
- the size of the cut/copied object may need to be altered to correspond to the scale of the objects in the target image.
- Object class information may comprise information about the relative size of objects of different classes. For example, objects of class building can be specified as larger by a certain factor than objects of class person. Using this information in conjunction with depth ordering information it is possible to automatically determine an appropriate size for the object to be pasted into the target image. In another example, the focus of the object to be pasted may need to be altered to match the focus in the target image.
- depth ordering information can be used together with information about depth of field information in the target image to decide what focus to apply to the object to be pasted.
- the colour of the object to be pasted can be altered according to colour details of the target image.
- Other options include changing the illumination or orientation of the object to be pasted, changing facial expressions in images of people to match those of objects in the target image and others.
- FIGs. 7a and 7b illustrate an example of using smart copy and paste.
- FIG. 7a schematically shows an image of a person 71 standing behind a post 70.
- the image of the person is to be copied and pasted into a reference image, referred to in this example as a target image 74 (see FIG 7b).
- the result of copying the image of the person is indicated in FIG 7a and it can be seen that a blank region 72 exists where the post 70 previously occluded part of the image of the person.
- the object-level information about the person is used which provides segmentation for the person and depth ordering information about the post and person.
- the occluded region 72 is filled in using tiles to yield a complete image 73 (see FIG. 7b). That complete image 73 is to be pasted into target image 74 which shows a house.
- target image 74 which shows a house.
- Object-level information about the target image indicates that the relative size of a house and a person is such that the complete image 73 needs to be resized. After appropriate re-sizing, and any other determined edits such as refocusing or recolouring the image of the person 75 is pasted into the reference image.
- FIG. 8 is an example of an output image produced from our object-level image editing system after performing improved digital tapestry.
- digital tapestry is used herein to refer to using parts of a plurality of source images and blending those parts together to form a single output image. In the example shown in FIG. 8 12 source images were used to form the single digital tapestry output image.
- any suitable method of blending the image parts together can be used.
- our earlier European Patent application number 05254745.2, filed on 28 July 2005 , entitled “Image Blending” describes a suitable method of digital tapestry.
- Our CVPR conference paper entitled “ Digital Tapestry” Carsten Rother, Sanjiv Kumar, Vladimir Kolmogorov, Andrew Blake, 2005, San Diego, USA also describes a suitable method of digital tapestry.
- digital tapestry is improved by enabling selection of image parts from the source image to be guided using the object-level information about the source images.
- selection of positions in the output image to place those image parts is guided using the object-level information.
- FIG. 9 is a flow diagram of a method of improved digital tapestry or image blending. It comprises accessing object-level information about each of a plurality of source images using the methods described above (see box 90).
- the object-level information is used to select source image region to be used (see box 91).
- object class information may be used in conjunction with specified criteria or rules to select pertinent image regions. If there are many images of people for example, it may be specified to select only a threshold number of these for inclusion. Criteria or rules may be specified to prefer some classes of object over others, for example, with grass or roads given lower preference than animals or people. It is also possible to give preference to objects of rare classes of which only one or two for example exist in the source images.
- the object-level information may also be used to automatically select locations in the output image for the selected source image regions (see box 92).
- source image regions may be positioned such that objects of class sky are placed near the top of the output image.
- the source image regions are then blended to form the digital tapestry as discussed above (see box 93).
- FIG. 8 it can be seen that the source image regions containing sky are all positioned so that the sky is at the top of the composite, output image.
- the images of people are selected such that different people are shown, or where the same person is shown more than once, different poses of that person are given.
- Super resolution is the process of making a high resolution image from a low resolution one.
- Existing super resolution methods ignore the type of object when creating the high resolution image and hence often fill in inappropriate details.
- An example of existing super resolution methods is given in William T. Freeman, Thouis R. Jones, and Egon C. Pasztor, "Example-based super-resolution", IEEE Computer Graphics and Applications, March/April, 2002 .
- FIG. 10 is a flow diagram of a method of using object-level information to improve super resolution.
- the source image has lower resolution than the output image.
- Object level information is accessed about the source image (see box 110) as described above.
- a database of patches or tiles of images having higher resolution than the source image is then accessed (see box 110).
- the image editing system selects patches or tiles of the same object class or object part as an object in the source image (see box 112) and constructs a high resolution image using the selected patches or tiles (see box 113).
- the tiles and the method of filling using those tiles may be as described above in the section headed "smart tiling" or any other suitable method may be used.
- a common need in image editing is the ability to crop images to create aesthetically pleasing images in which the proportions of various objects depicted and the relative positions of objects and colours are harmonious and pleasant. It is also required to be able to crop images to meet particular requirements, such as passport photographs, and to illustrate objects or scenes in the most clear and understandable way. Previously artists and graphic designers have manually cropped digital images to achieve these and similar objectives. However, there is a desire to enable non-skilled users to crop images effectively and to provide systems which automatically crop images without the need for human intervention.
- FIG. 11. is a flow diagram of a method of automatically cropping a source image using object-level information.
- the object level information about the source image is accessed (see box 120) as described above.
- the image editing system then accesses rules, criteria and/or templates (which may be probabilistic) for image cropping.
- rules, criteria and/or templates are pre-specified. For example, they are created on the basis of empirical data about how expert artists and image croppers transform a given digital image into a cropped one.
- the rules, criteria and/or templates may be associated with particular situations, such as templates for passport photographs, templates for landscape photographs, templates for portrait photographs, rules for images of still life objects, rules for images of moving objects etc.
- An example of a rule may be to preserve a ratio of 1 /3 sky to 2/3 ground in an image of a landscape scene.
- Appropriate ones of the rules, criteria and/or templates are then selected (see box 122). This selection can be made on the basis of user input or automatically on the basis of the object level information in the source image. For example, if the object level information indicates that the image comprises a person in the foreground and grass in the background, the portrait photograph template may be selected.
- the source image is then cropped (see box 123).
- FIG. 12 is a flow diagram of a method of automatically colour balancing an image using object -level information.
- the object -level information about the source image is accessed (see box 130) as described above.
- a comparison is then made (see box 131) of the colour details of an object in the source image with pre-specified colour details about objects of the same class (see box 131).
- pre-specified colour details about objects of the same class see box 131.
- the colour details compared can be colour distributions, colour values, particular colour statistics or any other suitable colour information. It is then possible to adjust the colour of the source image on the basis of the comparison (see box 132).
- the colour of only the object can be changed or of the whole image.
- object-level information in the image editing system as described herein is particularly advantageous because it enables single-click (or single user action) object selection. This ability is particularly useful for image editors who typically need to edit given objects in an image separately or independently of the rest of the image. Previously, it has been necessary to manually draw around required objects to select them, using lassoing tools, drawing tools or the like. This is time consuming and requires fine manual dexterity and expertise.
- the single-click object selection of the present application avoids these problems and is also referred to herein as TouchCut.
- FIG. 13 is a flow diagram of such a method.
- Object-level information is accessed about the source image (see box 140) as described above.
- the image editing system receives a user input selecting an object in the source image.
- This user input can be of any suitable type such as a mouse click, touch screen event, key press, sound or other user input.
- the user input requires only a single user-action to be made in order to select the object (see box 141). This is a huge improvement over the prior art where objects generally had to be selected by drawing round their borders.
- the image editing system selects the object (box 142) and may or may not indicate this via the user interface, for example, by highlighting the selected object, presenting a flashing border around the selected object or in other ways.
- FIG. 14 shows a source image 20 of a group of four cows against a background of grass. TouchCut is used to select the foreground object, being the group of four cows 22. This is achieved with a single mouse click anywhere in the region of the source image 20 comprising the cows.
- Once TouchCut is used to select an object that object can be manipulated in any suitable manner, for example to emphasise that object by changing its colour, focus or size, by adding borders to that object, adding shading, changing the depth ordering of the object relative to other objects in the image, or in any other suitable manner. Alternatively, such actions or manipulations can be made to the parts of the source image which are not selected.
- FIG. 1 5 shows example user interface displays of our image editing system where SmartReplace is used.
- a first user interface display 50 comprises an image of an aeroplane standing on grass in front of a building above which sky is visible.
- the sky object 54 has been selected using TouchCut or in any other suitable manner. In this example the selection is indicated by a dotted border around the sky object 54.
- a menu 53 of possible actions to be performed on the sky object is presented. The menu comprises the options: replace and colour levels in this example.
- a command bar is presented indicating that the replace function is current and providing two example sky object instances 55 to use to replace the existing sky. The user selects one of these object instances and clicks apply. The sky object of the first display 50 is then replaced by the selected sky object instance as indicated in the second display 51.
- the command bar now shows the option to erase the aeroplane object or to fill in again 56 and this can be done whilst ignoring the object labels 57 if required.
- FIG. 16 is a flow diagram of a method of SmartReplace.
- the image editing system accesses object-level information about that object and/or the source image as a whole (see box 170).
- a reference object is then selected from a database (see box 171).
- the reference object is preferably of the same class or a related class to the selected object. For example, clear sky may be replaced by a cloudy sky. Alternatively, an image of a cow which is unclear may be replaced by an image of a horse.
- the selection may also be made on the basis of object-level information about the source image as a whole. For example, that information might indicate that the source image is a landscape photograph so that object instances inappropriate for inclusion in such a landscape are not selected.
- the next stage involves replacing the selected object with the reference object (see box 172).
- the smart replace process comprises object synthesis. For example, suppose that it is required to change the image of a particular instance of a person by replacing the hair on that person. In this case the object-level information is used to identify and locate the hair object part of the person image. Tiles are picked from a database such that the tiles are from another instance of a hair object part. Those selected tiles are then used to replace the existing hair in the image. In this way a new object instance of a particular class is created; that new object instance not having been part of the database.
- the object-level information can also be used for smart depth of field manipulation.
- FIG. 17 shows two user interface displays 180, 181 from the image editing system.
- the first user interface display 180 shows a source image of a person 31 standing in the foreground with a grassed area receding into the background.
- TouchCut or any other suitable method is used to select the person 31 as described above.
- a menu 32 is presented indicating the actions that are possible with respect to the image of the person 31. These include for example, erase, focus, fix red eye and colour levels.
- focus is selected. This is indicated in a command bar in the second user interface display 181.
- a slider bar for adjusting the level of blur 33 is provided and when the amount of blur is selected to be high the results are shown in display 181. It can be seen that the image of the person 31 is retained in focus whilst blur is applied to the rest of the image. However, this is not essential.
- the depth of field can be adjusted in any suitable way taking into account the object-level information.
- the menu 32 associated with the selected image of a person 31 comprises options suitable for operation on the image of the person. These comprise: erase, focus, fix red eye, colour/levels.
- FIG. 1 5 shows menu 53 associated with the object, sky. This menu only presents the options of replace and colour/levels. This is because operations such as fix red eye and focus are not appropriate for objects of class sky.
- Menu options may depend on attributes of the object class, e.g. is it structured or does it belong to a broader category e.g. vehicles.
- FIG. 18 is a flow diagram of a method of providing a user interface of an image editing system using object level information.
- Object-level information is accessed about an image being displayed at the user interface (see box 190).
- An object in the image is then selected (see box 191). For example, this is achieved using a user input or is carried out automatically by the editing system.
- Class specific editing options are then accessed from a store of specified class specific editing options. Editing options appropriate for the class of the selected object are accessed (see box 192) and presented via the user interface (see box 193).
- a remote computer may store an example of the process described as software.
- a local or terminal computer may access the remote computer and download a part or all of the software to run the program.
- the local computer may download pieces of the software as needed, or execute some software instructions at the local terminal and some at the remote computer (or computer network).
- a dedicated circuit such as a DSP, programmable logic array, or the like.
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Processing Or Creating Images (AREA)
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
EP06270025.7A EP1826724B8 (fr) | 2006-02-28 | 2006-02-28 | Edition d'image au niveau objet utilisant les titres des données d'image |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
EP06270025.7A EP1826724B8 (fr) | 2006-02-28 | 2006-02-28 | Edition d'image au niveau objet utilisant les titres des données d'image |
Publications (3)
Publication Number | Publication Date |
---|---|
EP1826724A1 true EP1826724A1 (fr) | 2007-08-29 |
EP1826724B1 EP1826724B1 (fr) | 2015-03-25 |
EP1826724B8 EP1826724B8 (fr) | 2015-06-03 |
Family
ID=36720126
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
EP06270025.7A Not-in-force EP1826724B8 (fr) | 2006-02-28 | 2006-02-28 | Edition d'image au niveau objet utilisant les titres des données d'image |
Country Status (1)
Country | Link |
---|---|
EP (1) | EP1826724B8 (fr) |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US8515194B2 (en) | 2007-02-21 | 2013-08-20 | Microsoft Corporation | Signaling and uses of windowing information for images |
CN105608716A (zh) * | 2015-12-21 | 2016-05-25 | 联想(北京)有限公司 | 一种信息处理方法及电子设备 |
US9569697B1 (en) * | 2015-02-19 | 2017-02-14 | Google Inc. | Object oriented image editing |
CN111223118A (zh) * | 2018-11-27 | 2020-06-02 | 富士通株式会社 | 图像处理装置、图像处理方法及计算机可读记录介质 |
CN112446886A (zh) * | 2019-08-30 | 2021-03-05 | 上海商汤临港智能科技有限公司 | 图像处理方法及相关产品 |
Families Citing this family (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US12001642B2 (en) | 2021-04-19 | 2024-06-04 | Apple Inc. | User interfaces for managing visual content in media |
US11671696B2 (en) | 2021-04-19 | 2023-06-06 | Apple Inc. | User interfaces for managing visual content in media |
US20230319224A1 (en) * | 2022-03-10 | 2023-10-05 | Apple Inc. | User interfaces for managing visual content in a media representation |
Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20040165787A1 (en) | 2003-02-24 | 2004-08-26 | Patrick Perez | Image region filling by example-based tiling |
EP1748389A1 (fr) | 2005-07-28 | 2007-01-31 | Microsoft Corporation | Méthode et dispositif de melange des images |
-
2006
- 2006-02-28 EP EP06270025.7A patent/EP1826724B8/fr not_active Not-in-force
Patent Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20040165787A1 (en) | 2003-02-24 | 2004-08-26 | Patrick Perez | Image region filling by example-based tiling |
EP1748389A1 (fr) | 2005-07-28 | 2007-01-31 | Microsoft Corporation | Méthode et dispositif de melange des images |
Non-Patent Citations (10)
Title |
---|
BARRETT W A ET AL: "Object-based image editing", COMPUTER GRAPHICS PROCEEDINGS, PROCEEDINGS OF SIGGRAPH ANNUAL INTERNATIONAL CONFERENCE ON COMPUTER GRAPHICS AND INTERACTIVE TECHNIQUES, XX, XX, 2002, pages 1 - 8, XP002338402 * |
BOLDYS J: "Bayesian supervised segmentation of objects in natural images using low-level information", IMAGE AND SIGNAL PROCESSING AND ANALYSIS, 2003. ISPA 2003. PROCEEDINGS OF THE 3RD INTERNATIONAL SYMPOSIUM ON ROME, ITALY SEPT. 18-20, 2003, PISCATAWAY, NJ, USA,IEEE, vol. 2, 18 September 2003 (2003-09-18), pages 1054 - 1059, XP010704013, ISBN: 953-184-061-X * |
CARSTEN ROTHER ET AL., DIGITAL TAPESTRY, 2005 |
CHEN H ET AL ASSOCIATION FOR COMPUTING MACHINERY: "EXAMPLE-BASED COMPOSITE SKETCHING OF HUMAN PORTRAITS", PROCEEDINGS NPAR 2004. 3RD. INTERNATIONAL SYMPOSIUM ON NON-PHOTOREALISTIC ANIMATION AND RENDERING. ANNECY, FRANCE, JUNE 7 - 9, 2004, SYMPOSIUM ON NON - PHOTOREALISTIC ANIMATION AND RENDERING, NEW YORK, NY : ACM, US, 7 June 2004 (2004-06-07), pages 95 - 102, XP001210027, ISBN: 1-58113-887-3 * |
NICHOLAS DIAKOPOULOS ET AL: "Content Based Image Synthesis", CONFERENCE ON IMAGE AND VIDEO RETRIEVAL (CIVR) 2004, July 2004 (2004-07-01), Dublin, ireland, pages 299 - 307, XP019008942 * |
RYTSAR ET AL.: "interactive segmentation with hidden objec-based annotations: towards smart media", PROCEEDINGS OF THE SPIE ELECTRONICS IMAGING, STORAGE AND RETRIEVAL METHODS AND APPLICATIONS FOR MULTIMEDIA, January 2004 (2004-01-01), San Jose, CA, USA, pages 1 - 9, XP002393102 * |
WILLIAM T. FREEMAN; THOUIS R. JONES; EGON C. PASZTOR: "Example-based super- resolution", IEEE COMPUTER GRAPHICS AND APPLICATIONS, March 2002 (2002-03-01) |
WINN J ET AL: "Object Categorization by Learned Universal Visual Dictionary", COMPUTER VISION, 2005. ICCV 2005. TENTH IEEE INTERNATIONAL CONFERENCE ON BEIJING, CHINA 17-20 OCT. 2005, PISCATAWAY, NJ, USA,IEEE, 17 October 2005 (2005-10-17), pages 1800 - 1807, XP010857031, ISBN: 0-7695-2334-X * |
WU C ET AL: "AUTOMATIC EYEGLASSES REMOVAL FROM FACE IMAGES", IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, IEEE SERVICE CENTER, LOS ALAMITOS, CA, US, vol. 26, no. 3, March 2004 (2004-03-01), pages 322 - 336, XP001190677, ISSN: 0162-8828 * |
YAOWU XU ET AL: "Object Segmentation and Labelingby Learning From Examples", IEEE TRANSACTIONS ON IMAGE PROCESSING, IEEE SERVICE CENTER, PISCATAWAY, NJ, US, vol. 12, no. 6, June 2003 (2003-06-01), XP011074408, ISSN: 1057-7149 * |
Cited By (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US8515194B2 (en) | 2007-02-21 | 2013-08-20 | Microsoft Corporation | Signaling and uses of windowing information for images |
US9123089B2 (en) | 2007-02-21 | 2015-09-01 | Microsoft Technology Licensing, Llc | Signaling and uses of windowing information for images |
US9569697B1 (en) * | 2015-02-19 | 2017-02-14 | Google Inc. | Object oriented image editing |
US10198820B1 (en) | 2015-02-19 | 2019-02-05 | Google Llc | Object oriented image editing |
US10872416B2 (en) | 2015-02-19 | 2020-12-22 | Google Llc | Object oriented image editing |
CN105608716A (zh) * | 2015-12-21 | 2016-05-25 | 联想(北京)有限公司 | 一种信息处理方法及电子设备 |
CN105608716B (zh) * | 2015-12-21 | 2020-12-18 | 联想(北京)有限公司 | 一种信息处理方法及电子设备 |
CN111223118A (zh) * | 2018-11-27 | 2020-06-02 | 富士通株式会社 | 图像处理装置、图像处理方法及计算机可读记录介质 |
CN112446886A (zh) * | 2019-08-30 | 2021-03-05 | 上海商汤临港智能科技有限公司 | 图像处理方法及相关产品 |
Also Published As
Publication number | Publication date |
---|---|
EP1826724B1 (fr) | 2015-03-25 |
EP1826724B8 (fr) | 2015-06-03 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
EP1826723B1 (fr) | Edition d'image au niveau objet | |
EP1826724B1 (fr) | Edition d'image au niveau objet utilisant les titres des données d'image | |
US8644644B2 (en) | Methods and apparatus for blending images | |
US6009209A (en) | Automated removal of red eye effect from a digital image | |
JP4398726B2 (ja) | 1つまたは複数の画像の自動フレーム選択およびレイアウトならびにフレームによって境界付けされた画像の生成 | |
US8594428B2 (en) | Interactive segmentation of images with single scribbles | |
EP1372109B1 (fr) | Procédé et système pour l'amélioration d'images portrait | |
JP3189870B2 (ja) | 画像処理装置 | |
US20040164996A1 (en) | Image region filling by exemplar-based inpainting | |
JP5299173B2 (ja) | 画像処理装置および画像処理方法、並びにプログラム | |
US11386589B2 (en) | Method and device for image generation and colorization | |
JPH11328380A (ja) | 画像処理装置、画像処理方法およびその方法をコンピュ―タに実行させるプログラムを記録したコンピュ―タ読み取り可能な記録媒体 | |
CN101606179B (zh) | 用于遮罩、选择和路径的通用前端 | |
CN101119430A (zh) | 模板生成装置、图像排版装置、修正的模板生成装置及其程序 | |
WO2013141144A1 (fr) | Appareil de traitement d'image, procédé de traitement d'image et support d'enregistrement lisible par ordinateur | |
CN116580127B (zh) | 图像生成方法、装置、电子设备和计算机可读存储介质 | |
CN106447756B (zh) | 用于生成用户定制的计算机生成动画的方法和系统 | |
Dayley et al. | Photoshop CS5 Bible | |
JP2011002875A (ja) | 描画支援装置,描画支援方法および描画支援プログラム | |
CN112637633B (zh) | 节目生成方法和节目生成装置 | |
Team et al. | Adobe Photoshop CC | |
KR102606373B1 (ko) | 영상에서 검출되는 얼굴의 랜드마크를 조절하기 위한 방법 및 장치 | |
CN117112826A (zh) | 一种图像生成方法、装置、计算机设备及存储介质 | |
CN117132693A (zh) | 场景渲染方法及装置 | |
Segovia et al. | Aesthetic role of transparency and layering in the creation of photo layouts |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PUAI | Public reference made under article 153(3) epc to a published international application that has entered the european phase |
Free format text: ORIGINAL CODE: 0009012 |
|
AK | Designated contracting states |
Kind code of ref document: A1 Designated state(s): AT BE BG CH CY CZ DE DK EE ES FI FR GB GR HU IE IS IT LI LT LU LV MC NL PL PT RO SE SI SK TR |
|
AX | Request for extension of the european patent |
Extension state: AL BA HR MK YU |
|
17P | Request for examination filed |
Effective date: 20080218 |
|
17Q | First examination report despatched |
Effective date: 20080331 |
|
AKX | Designation fees paid |
Designated state(s): AT BE BG CH CY CZ DE DK EE ES FI FR GB GR HU IE IS IT LI LT LU LV MC NL PL PT RO SE SI SK TR |
|
GRAP | Despatch of communication of intention to grant a patent |
Free format text: ORIGINAL CODE: EPIDOSNIGR1 |
|
INTG | Intention to grant announced |
Effective date: 20141006 |
|
GRAS | Grant fee paid |
Free format text: ORIGINAL CODE: EPIDOSNIGR3 |
|
GRAA | (expected) grant |
Free format text: ORIGINAL CODE: 0009210 |
|
AK | Designated contracting states |
Kind code of ref document: B1 Designated state(s): AT BE BG CH CY CZ DE DK EE ES FI FR GB GR HU IE IS IT LI LT LU LV MC NL PL PT RO SE SI SK TR |
|
REG | Reference to a national code |
Ref country code: GB Ref legal event code: FG4D |
|
REG | Reference to a national code |
Ref country code: CH Ref legal event code: EP Ref country code: CH Ref legal event code: PUE Owner name: MICROSOFT TECHNOLOGY LICENSING, LLC, US Free format text: FORMER OWNER: MICROSOFT CORPORATION, US |
|
RAP2 | Party data changed (patent owner data changed or rights of a patent transferred) |
Owner name: MICROSOFT TECHNOLOGY LICENSING, LLC |
|
REG | Reference to a national code |
Ref country code: IE Ref legal event code: FG4D |
|
REG | Reference to a national code |
Ref country code: DE Ref legal event code: R096 Ref document number: 602006044885 Country of ref document: DE Effective date: 20150430 |
|
REG | Reference to a national code |
Ref country code: AT Ref legal event code: REF Ref document number: 718257 Country of ref document: AT Kind code of ref document: T Effective date: 20150515 |
|
PG25 | Lapsed in a contracting state [announced via postgrant information from national office to epo] |
Ref country code: FI Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT Effective date: 20150325 Ref country code: SE Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT Effective date: 20150325 Ref country code: LT Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT Effective date: 20150325 |
|
REG | Reference to a national code |
Ref country code: AT Ref legal event code: MK05 Ref document number: 718257 Country of ref document: AT Kind code of ref document: T Effective date: 20150325 |
|
REG | Reference to a national code |
Ref country code: LT Ref legal event code: MG4D |
|
PG25 | Lapsed in a contracting state [announced via postgrant information from national office to epo] |
Ref country code: GR Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT Effective date: 20150626 Ref country code: LV Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT Effective date: 20150325 |
|
PG25 | Lapsed in a contracting state [announced via postgrant information from national office to epo] |
Ref country code: NL Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT Effective date: 20150325 |
|
PG25 | Lapsed in a contracting state [announced via postgrant information from national office to epo] |
Ref country code: CZ Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT Effective date: 20150325 Ref country code: ES Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT Effective date: 20150325 Ref country code: EE Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT Effective date: 20150325 Ref country code: SK Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT Effective date: 20150325 Ref country code: PT Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT Effective date: 20150727 Ref country code: RO Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT Effective date: 20150325 |
|
PG25 | Lapsed in a contracting state [announced via postgrant information from national office to epo] |
Ref country code: IS Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT Effective date: 20150725 Ref country code: AT Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT Effective date: 20150325 Ref country code: PL Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT Effective date: 20150325 |
|
REG | Reference to a national code |
Ref country code: DE Ref legal event code: R097 Ref document number: 602006044885 Country of ref document: DE |
|
REG | Reference to a national code |
Ref country code: FR Ref legal event code: PLFP Year of fee payment: 11 |
|
PG25 | Lapsed in a contracting state [announced via postgrant information from national office to epo] |
Ref country code: DK Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT Effective date: 20150325 |
|
PLBE | No opposition filed within time limit |
Free format text: ORIGINAL CODE: 0009261 |
|
STAA | Information on the status of an ep patent application or granted ep patent |
Free format text: STATUS: NO OPPOSITION FILED WITHIN TIME LIMIT |
|
26N | No opposition filed |
Effective date: 20160105 |
|
PG25 | Lapsed in a contracting state [announced via postgrant information from national office to epo] |
Ref country code: IT Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT Effective date: 20150325 |
|
PG25 | Lapsed in a contracting state [announced via postgrant information from national office to epo] |
Ref country code: BE Free format text: LAPSE BECAUSE OF NON-PAYMENT OF DUE FEES Effective date: 20160229 Ref country code: SI Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT Effective date: 20150325 |
|
PG25 | Lapsed in a contracting state [announced via postgrant information from national office to epo] |
Ref country code: BE Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT Effective date: 20150325 |
|
PG25 | Lapsed in a contracting state [announced via postgrant information from national office to epo] |
Ref country code: LU Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT Effective date: 20160228 Ref country code: MC Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT Effective date: 20150325 |
|
REG | Reference to a national code |
Ref country code: CH Ref legal event code: PL |
|
PG25 | Lapsed in a contracting state [announced via postgrant information from national office to epo] |
Ref country code: CH Free format text: LAPSE BECAUSE OF NON-PAYMENT OF DUE FEES Effective date: 20160229 Ref country code: LI Free format text: LAPSE BECAUSE OF NON-PAYMENT OF DUE FEES Effective date: 20160229 |
|
REG | Reference to a national code |
Ref country code: IE Ref legal event code: MM4A |
|
REG | Reference to a national code |
Ref country code: FR Ref legal event code: PLFP Year of fee payment: 12 |
|
PG25 | Lapsed in a contracting state [announced via postgrant information from national office to epo] |
Ref country code: IE Free format text: LAPSE BECAUSE OF NON-PAYMENT OF DUE FEES Effective date: 20160228 |
|
REG | Reference to a national code |
Ref country code: FR Ref legal event code: PLFP Year of fee payment: 13 |
|
PG25 | Lapsed in a contracting state [announced via postgrant information from national office to epo] |
Ref country code: CY Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT Effective date: 20150325 Ref country code: HU Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT; INVALID AB INITIO Effective date: 20060228 |
|
PG25 | Lapsed in a contracting state [announced via postgrant information from national office to epo] |
Ref country code: TR Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT Effective date: 20150325 |
|
PG25 | Lapsed in a contracting state [announced via postgrant information from national office to epo] |
Ref country code: BG Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT Effective date: 20150325 |
|
REG | Reference to a national code |
Ref country code: DE Ref legal event code: R082 Ref document number: 602006044885 Country of ref document: DE |
|
PGFP | Annual fee paid to national office [announced via postgrant information from national office to epo] |
Ref country code: GB Payment date: 20220106 Year of fee payment: 17 Ref country code: DE Payment date: 20220105 Year of fee payment: 17 |
|
PGFP | Annual fee paid to national office [announced via postgrant information from national office to epo] |
Ref country code: FR Payment date: 20220118 Year of fee payment: 17 |
|
P01 | Opt-out of the competence of the unified patent court (upc) registered |
Effective date: 20230501 |
|
REG | Reference to a national code |
Ref country code: DE Ref legal event code: R119 Ref document number: 602006044885 Country of ref document: DE |
|
GBPC | Gb: european patent ceased through non-payment of renewal fee |
Effective date: 20230228 |
|
PG25 | Lapsed in a contracting state [announced via postgrant information from national office to epo] |
Ref country code: GB Free format text: LAPSE BECAUSE OF NON-PAYMENT OF DUE FEES Effective date: 20230228 |
|
PG25 | Lapsed in a contracting state [announced via postgrant information from national office to epo] |
Ref country code: GB Free format text: LAPSE BECAUSE OF NON-PAYMENT OF DUE FEES Effective date: 20230228 Ref country code: FR Free format text: LAPSE BECAUSE OF NON-PAYMENT OF DUE FEES Effective date: 20230228 Ref country code: DE Free format text: LAPSE BECAUSE OF NON-PAYMENT OF DUE FEES Effective date: 20230901 |